• 제목/요약/키워드: Contents Research

검색결과 18,783건 처리시간 0.044초

Restricting Answer Candidates Based on Taxonomic Relatedness of Integrated Lexical Knowledge Base in Question Answering

  • Heo, Jeong;Lee, Hyung-Jik;Wang, Ji-Hyun;Bae, Yong-Jin;Kim, Hyun-Ki;Ock, Cheol-Young
    • ETRI Journal
    • /
    • 제39권2호
    • /
    • pp.191-201
    • /
    • 2017
  • This paper proposes an approach using taxonomic relatedness for answer-type recognition and type coercion in a question-answering system. We introduce a question analysis method for a lexical answer type (LAT) and semantic answer type (SAT) and describe the construction of a taxonomy linking them. We also analyze the effectiveness of type coercion based on the taxonomic relatedness of both ATs. Compared with the rule-based approach of IBM's Watson, our LAT detector, which combines rule-based and machine-learning approaches, achieves an 11.04% recall improvement without a sharp decline in precision. Our SAT classifier with a relatedness-based validation method achieves a precision of 73.55%. For type coercion using the taxonomic relatedness between both ATs and answer candidates, we construct an answer-type taxonomy that has a semantic relationship between the two ATs. In this paper, we introduce how to link heterogeneous lexical knowledge bases. We propose three strategies for type coercion based on the relatedness between the two ATs and answer candidates in this taxonomy. Finally, we demonstrate that this combination of individual type coercion creates a synergistic effect.

Vision-based garbage dumping action detection for real-world surveillance platform

  • Yun, Kimin;Kwon, Yongjin;Oh, Sungchan;Moon, Jinyoung;Park, Jongyoul
    • ETRI Journal
    • /
    • 제41권4호
    • /
    • pp.494-505
    • /
    • 2019
  • In this paper, we propose a new framework for detecting the unauthorized dumping of garbage in real-world surveillance camera. Although several action/behavior recognition methods have been investigated, these studies are hardly applicable to real-world scenarios because they are mainly focused on well-refined datasets. Because the dumping actions in the real-world take a variety of forms, building a new method to disclose the actions instead of exploiting previous approaches is a better strategy. We detected the dumping action by the change in relation between a person and the object being held by them. To find the person-held object of indefinite form, we used a background subtraction algorithm and human joint estimation. The person-held object was then tracked and the relation model between the joints and objects was built. Finally, the dumping action was detected through the voting-based decision module. In the experiments, we show the effectiveness of the proposed method by testing on real-world videos containing various dumping actions. In addition, the proposed framework is implemented in a real-time monitoring system through a fast online algorithm.

Automatic Container Code Recognition from Multiple Views

  • Yoon, Youngwoo;Ban, Kyu-Dae;Yoon, Hosub;Kim, Jaehong
    • ETRI Journal
    • /
    • 제38권4호
    • /
    • pp.767-775
    • /
    • 2016
  • Automatic container code recognition from a captured image is used for tracking and monitoring containers, but often fails when the code is not captured clearly. In this paper, we increase the accuracy of container code recognition using multiple views. A character-level integration method combines recognized codes from different single views to generate a new code. A decision-level integration selects the most probable results from the codes from single views and the new integrated code. The experiment confirmed that the proposed integration works successfully. The recognition from single views achieved an accuracy of around 70% for the test images collected on a working pier, whereas the proposed integration method showed an accuracy of 96%.

쌍화탕과 발효쌍화탕 분획물의 폴리페놀함량 및 항산화 활성 (Polyphenol Contents and Antioxidant Activities of Fractions from Ssanghwa-tang and Fermented Ssanghwa-tang)

  • 김동선;엄영란;양민철;윤나영;마진열
    • 한국한의학연구원논문집
    • /
    • 제16권3호
    • /
    • pp.175-178
    • /
    • 2010
  • The aim of this study is to compare antioxidant activity and total polyphenol contents between before and after fermentation of Ssanghwa-tang according to solvent partition. Ssanghwa-tang was fermented with Lactobacilus fermentum. Ssanghwa-tang and the fermented Ssanghwa-tang were fractioned by solvent partition with ethyl acetate, butanol and water. The Ssanghwa-tangs and their solvent fractions were evaluated for total polyphenol contents and DPPH radical scavenging activity. The antioxidant activity as well as the total polyphenol contents were highest in each ethyl acetate fraction and significantly (p<0.05) increased after fermentation.

Blackbox and Scenario-Based Testing of Online Games Using Game Description Language

  • Cho, Chang-Sik;Lee, Dong-Chun;Sohn, Kang-Min;Park, Chang-Joon;Kang, Ji-Hoon
    • ETRI Journal
    • /
    • 제33권3호
    • /
    • pp.470-473
    • /
    • 2011
  • In this letter, we propose blackbox and scenario-based testing of multiplayer online games as well as simple load testing. Game testing is done from outside the source code, and the access to the source code is not required to testers because the game logic is described with a game description language and virtual game map. Instead of using a subset of the main game client for the test client, only game packet protocols and the sequence of packets are analyzed for new game testing. In addition, complex and various scenarios can be tested through combining defined actions. Scenario-based testing helps testers mimic real testing environments instead of doing simple load testing and improves the productivity of game testing.

A Novel Method for Bitrate Control within Macroblocks Using Kalman and FIR Filters

  • Seok, Jin-Wuk;Yoon, Ki-Song;Kim, Bum-Ho;Lee, Jeong-Woo
    • ETRI Journal
    • /
    • 제33권4호
    • /
    • pp.641-644
    • /
    • 2011
  • In this letter, we propose a novel bitrate control, using both Kalman and FIR filters, based on a Hamiltonian analysis with respect to the amount of bits from each macroblock, in an encoding of a general video codec such as H.264/AVC. Since the proposed bitrate control is based on the simple computation of an optimal control method based on the Hamiltonian analysis, it is not necessary to use additional computation, such as a DCT or quantization, to estimate the bits for bitrate control. As a result, the proposed algorithm can be applied to single-pass encoding and can provide sufficient encoding speed with respect to various applications, even those requiring real-time control.